Casting your own parts in resin

[dogfish7 (R.I.P.)]

Looking very professional Beekster.

[Beekster]

May 2, 2020 8:37:24 GMT -6 dogfish7 (R.I.P.) said:

Looking very professional Beekster.

Thank you!  With home casting you can achieve pretty professional results, at least for simpler parts and molds.  The real pros use harder RTV, which lasts longer so you get more pours before it wears out.  They also use vacuum systems to suck all the air bubbles out as the RTV cures, to ensure that all the detail is reproduced faithfully in the mold.  Similarly, they usually have either vacuum or pressure systems for the resin pours too, so that air bubbles in the resin are sucked/forced out of the mold before it cures.  It follows that you need a pot life for the resin that is much longer than the few minutes available with this particular Alumilite material, so that the pump can evacuate or pressurize the chamber and get the bubbles out before the resin begins to harden in the mold.  All this fancy stuff is necessary for doing large castings, like Sherman tank hulls or turrets and still larger things like the trailer beds for Accurate Armour's tank transporter kits.  Something else to consider with larger objects is warping of the casting, which obviously we don't want.  Stiffer RTV is part of the solution, as is using more of it so the mold itself is beefier.  With the kind of stuff I'm doing, I don't worry too much about warping in the castings; there isn't enough mass of resin to cause distortion.

As of an hour ago, I've used up the last of this initial batch of RTV and poured a couple of more molds.  I still have one or two do do for pieces not yet molded, and I will make new molds for a couple of the parts on that gang mold that have been giving me problems.  A couple of things will probably require a two-piece mold.  I've done those before, but ages ago and they do tend to create a lot of flash on the finished castings.  I will have to devote some time to engineering those better than I have done before.  I do still have some old ones around to look at, and get the thought process moving.  And as this goes on, I will address what to do with failed castings; just because air bubbles cause a flaw doesn't mean you can't fix things.  While the chemistry here is affordable (I'm into this now $65, and still have three quarters of the resin chemicals left to do more pours), and more affordable than stocking up on Academy TD kits just to use their interiors, you still don't want to throw away a casting if a little more effort will make it usable.  Stay tuned...

[Dukemaddog]

There may be flaws, but they are not visible with those parts in place. Nice work!

[Beekster]

May 2, 2020 11:32:43 GMT -6 Dukemaddog said:

There may be flaws, but they are not visible with those parts in place. Nice work!

The parts shown indeed have quite small flaws.  Others have more serious problems, but that just provides material to illustrate how to fix short-shot casting problems.  I've been working all morning with epoxy putty, under magnification, to fill in problem areas on several parts.  I have to leave everything alone now for that to set up, but I'll take some pictures of these alongside the plastic masters to illustrate how to fix stuff.  Then more pictures once I have finished with the file work necessary to refine the shapes.

[Beekster]

As promised, step one of fixing flaws. Simple surface bubbles can often be fixed with a drop of superglue, with the surface filed flat afterwards. But missing sections caused by bubbles require reconstructive surgery. Take a look at the images below:



You can see where missing contours caused by air bubbles are filled in and roughly shaped in epoxy putty, and in one instance a bit of embedded styrene rod. Once the epoxy putty has cured, the proper contours can be filed as required. A few bolt heads are missing, and these two will be easy to replace once the structural work is done. These shots were taken under high magnification; for reference the master pattern is about five eighths of an inch long and tall, and half an inch across.

Needless to say, this same kind of repair can be done on resin kits too, and may be the only way of dealing with problems for an out of production kit.

[JED]

Very informative and great results

[Beekster]

A setback this morning. One of the molds I made yesterday failed to cure properly, and I will have to scrub the remains off the masters and mount them again for another try when the next batch of RTV arrives. The other one will probably be workable, but will need another couple of days to continue to cure. Time will tell...

Why did this happen? I pulled apart the mold boxes a couple of hours ago, and the material had been curing for nearly 22 hours at that point and the top surfaces were solid. It's possible I should have left them alone for another few hours, but a solid top surface is generally indicative that you can dismantle the box and pull the masters out. It's also possible that the last bits of the vulcanizing chemical were more carrier fluid than active ingredients; getting the material thoroughly mixed isn't guaranteed when the quantity in the bottle is so small, as it is with the starter kit. That's probably why they want you to use it all at once, but if you go that route you had better have a whole bunch of molds boxes & masters put together just waiting for one big pour of RTV. That isn't impossible, but beware of the bubbles: You probably don't have enough hands to tap all the poured molds on a hard surface for a couple of minutes to get the air bubbles jiggled out. Not unless you happen to be a Hindu deity; I most certainly am not.

[dogfish7 (R.I.P.)]

If the A and B mixes are not Exactly the same amounts, this can happen. Had the same thing happen to me. One will need a compression Pot as well to drive all the air out.

[Beekster]

May 3, 2020 10:04:57 GMT -6 dogfish7 (R.I.P.) said:

If the A and B mixes are not Exactly the same amounts, this can happen. Had the same thing happen to me. One will need a compression Pot as well to drive all the air out.

Precisely.  And with a ten to one ratio for mixing the RTV, delineated by mass in grams in the instructions but estimated by liquid volume in milliliters in practice, it's not difficult to get things a bit off.  This is why I usually try to err of the side of a bit more vulcanizer than the required amount, since more vulcanizer just speeds the curing process and ensures that it works.  The downside is that you will wind up with some RTV that you can't use, since you have run out of the vulcanizer.  I tried to avoid this, but didn't quite succeed.

The resin itself is much easier to deal with since it's equal amounts, easily managed with the 3ml medicine droppers, and a slight mismatch in quantity isn't a big problem.  A bit more hardener, and the resin sets up faster and generates more heat as it cures.  A bit less, and it takes longer to harden but eventually does.

[Beekster]

This is what I mean by restoring incomplete castings to make them whole. While there's a bit more work do do on these, not least replacing a few bolt heads that have decided to part ways with their brethren, enough is done to illustrate the scope of work. There's epoxy putty, superglue, styrene strip, and file & sand work to restore contours that were missing from air bubbles. A couple of tiny brown plastic bolt heads, and larger ones (.7 and .85mm) punched from plastic sheet. Once primed and painted, no one will be able to tell these apart from the kit pieces.

[JED]

Great job Beekster

[Beekster]

My shipment of RTV arrived day before yesterday. Yesterday morning I mixed and poured two molds, and both of those have cured and look good. I will do the first castings in them this afternoon. Two more mold boxes are ready for RTV, and that too will happen later today. I have changed the way I position some parts in hopes of making the new molds more reliable, so we'll see how that goes. There are also a few parts which I think will benefit from being done with two-part molds. A two-part mold is seen in process below:

The part is roughly half embedded in clay, and the pour stub is up against one wall of the mold box. Several round depressions have been pressed into the clay. When the box is finished and RTV is poured, the RTV will fill those depressions. The finished mold half will show raised stubs, which will serve as keys to properly align the other half of the mold. Naturally, this first pour must cure completely before the second half is poured. More on that when I get to that stage; there is a minor complication inherent to two-piece molds that we must deal with...

[Beekster]

Some partial successes, and some more failures. I have good parts for the chairs, but the mold broke pulling the castings out so that requires another try sometime to get a fixture pattern that doesn't put such stress on the RTV. I have what I need for now, so that will happen sometime in the future. I have good castings of the hull side mounts for the seats, some junction boxes, and a corner bin for the co-driver's compartment. The clutch and brake pedals for the driver won't cast regardless of orientation, so I will make those from styrene. I also haven't gotten the chair lift mechanisms to cast well, and will probably have to do those from scratch as well to preserve the originals for another try at molding later. I will hold off on that transmission for now, and make my first two-piece mold for the transmission shift linkage which is a complicated shape.

[dogfish7 (R.I.P.)]

Nice work! Do you think you may sell a few copies later?

[Beekster]

May 15, 2020 14:32:45 GMT -6 dogfish7 (R.I.P.) said:

Nice work! Do you think you may sell a few copies later?

No.  As most of these are direct copies of kit parts, there are intellectual property issues that I wouldn't want to run afoul of.  These aren't quite like the Chilean Sherman conversions I did several years ago, where I heavily modified any kit pieces I used so there was considerable wiggle room from a legal standpoint...and I didn't do the casting for those, I had a professional do that job; unfortunately he's no longer able to cast in commercial quantities which is why my kits are no longer available.

[dogfish7 (R.I.P.)]

Forgot about the Copy right laws. Makes perfect sense.

[JED]

Nice and interesting work Beekster

[Beekster]

And this is the transmission linkage in the first mold half, just now taken out of the Lego box:



Everything worked out as expected, and the part both comes out of this half easily and more importantly goes back in snugly. Now it's time to prepare for the next step:




The mold box is nearly assembled again, ready for the next pour. Sharp eyes will note a sheen on the surface of the mold, compared to the previous photo. That's because I have coated the mold surface with Vaseline, which will provide a barrier to prevent the next pour or RTV from sticking to the first one. All those raised blobs are keys, which will serve to keep the mold halves aligned and tightly together when resin is poured so the part isn't misshapen when it is cast.

[Beekster]

And this is the completed two=part mold:

The two pieces came apart easily, so I had enough Vaseline on there as a mold release agent. While everything was cured hard enough to disassemble, according to the RTV specifications this still has several hours to cure yet so I won't do a first casting until tomorrow morning. When I do that, it will be messy, as two-piece molds tend to be. I will dust a little baby powder onto both mold halves to draw the resin into the nooks and crannies, and when I mix the resin I will use a dropper to add a little resin to each half while sitting open like this, to try to minimize air bubbles. It's bound to run out the pour opening. I will have to quickly put the mold together and pour the rest of the resin, flexing and prodding to get air bubbles out. Some resin will inevitably get squeezed out the sides of the mold as I do this. There is usually a lot of resin flash on the casting when a two-piece mold is used, so clean-up can be tedious. But this is usually the only way to successfully cast intricate, odd, and/or unusual shapes like this shift linkage.

[JCON]

Ineteresting!!!